Portals
A US Regional Bank reinvents collections using AI/ML

case study

A US Regional Bank reinvents collections using AI/ML

A leading US Regional Bank offers retail and commercial banking products and services to individuals, small businesses and commercial industries. The US Regional Bank recognized the need to continue strengthening their customer-centric culture and financial discipline to achieve significant growth while maintaining a high standard of excellence.

This objective required leveraging advanced analytics to drive insight, advice, and tailored solutions. It represented the US Regional Bank’s foray into an untrusted AI/Machine Learning territory, which involved using smart analytics to develop trusted and proven models. The US Regional Bank identified its collections function as the first area to target to reap the benefits of this solution.

Key Challenges

  • Collections has a high cost due to the expense of call-based customer outreach for overdue load remediation, without a guaranteed return
  • Improving customer value while minimizing risk and meeting regulatory requirements

Ready to experience?

TALK TO EXPERTS
Line

The Solution

Business-led success with AI/ML enabled analytics

  • Cultivating the US Regional Bank’s trust by starting with a POC to address a specific business challenge with measurable success criteria before expanding
  • Utilized a pre-built data model and automated pipeline with smart search to expedite the Machine Learning training period
  • Used FinXEdge Collect explainability feature over a four-month risk validation process to ensure the model was robust and met regulatory compliance

Improving collections with advanced analytics

  • FinXEdge Collect leverages advanced ML to combine customer behavior pattern, available credit, employment, financial and transactional data to improve the collections process
  • Accurately predict risk scores for delinquent accounts by segmentation
  • Optimize customer contact by suggesting “prime-time” calling
Line

Benefits

Exceeded roll rate lift objective by more than 100%

Exceeded roll rate lift objective by more than 100%

Delivered expense savings with a reduction in staffing as a result of refined delay queuing and contact strategies

Delivered expense savings with a reduction in staffing as a result of refined delay queuing and contact strategies

Executed a credible business-led use case that supports expansion to broader data sets and collection functions

Executed a credible business-led use case that supports expansion to broader data sets and collection functions

Cultivation of trust and a strong belief in the potential of Machine Learning-based models, which are delivering benefits for collections and evaluate for more extensive use

Cultivation of trust and a strong belief in the potential of Machine Learning-based models, which are delivering benefits for collections and evaluate for more extensive use